We help companies in the US and Europe hire experienced engineers, from funded startups to teams like Klarna, Retool, and Pinterest. We’ve been doing this since 2015, and more than 1,500 companies have hired engineers through us.
About the client
A research-driven, AI-native cloud company building next-generation AI infrastructure. AI teams use its platform to run and scale AI workloads on large GPU clusters, and its own team has contributed to leading open-source research, models, and datasets. The goal is to sharply lower the cost of modern AI. The fleet is growing fast, and the work is automating it.
The role
As a Senior Software Engineer, you’d build the control plane behind that growth: durable provisioning workflows, self-healing infrastructure, and APIs that let the inference team create, scale, or retire clusters without filing tickets. This is production software development, with ownership from design through operation.
What you’ll do
- Build the provisioning state machine: model a host’s full lifecycle (discovery, bring-up, GPU driver/CUDA stack, health validation, decommission/RMA) as explicit, versioned states and transitions
- Build inference-ready Kubernetes clusters, custom control planes and scheduling/orchestration workflows for training, tuning and serving workloads
- Design declarative, manifest-driven APIs that let teams create, scale or retire clusters with one call
- Automate self-healing: detect failed nodes, drain them safely, trigger repair, and return healthy capacity to the pool
- Own pipeline reliability: idempotency, retries, rollback, and drift detection
Must-have requirements
- 6+ years of commercial software engineering experience
- 4+ years of production experience with Go: you write and test real software for a living
- 3+ years of hands-on Kubernetes experience, beyond just deploying apps
- Product mindset: you’ve built internal platforms or APIs used by other engineering teams and care about their developer experience
- Track record of shipping production code on substantial initiatives, not interrupt-driven support work
- High ownership and the ability to explain the systems you’ve built end-to-end
- Comfortable with AI-assisted coding
- Full professional English proficiency
Big plus
- Temporal/Cadence (durable, long-running workflows) or Kubernetes controllers/operators written in Go
- Event-driven systems: Kafka, NATS, SQS, pub/sub
- Bare-metal provisioning (PXE/iPXE, Redfish/IPMI, BMC) and networking fundamentals (VLANs, BGP, fabric design)
- GPU cluster stacks: NCCL, CUDA, InfiniBand/RoCE
- Systems programming in Rust or Go
Team and setup
- A growing, internationally distributed team of 8–9 engineers building inference-cluster infrastructure
- Close work with GPU systems and ML workloads: you’ll gain deep technical expertise, learn from strong senior engineers, and take increasing ownership of impactful infrastructure initiatives
- Fast-moving environment with direct collaboration with principal engineers and team leads
- Communication via Slack and video calls
Selection process
- Short form with a few questions about your experience
- 40-min intro call with a recruiter
- 30-min algorithmic task interview
- 1.5-hour full technical round
- Interviews with the client: coding interview and system design interview
